709. Risk Factors for Candida auris Candidemia: Results from a Multicenter Case-Control Study

Abstract Background The emergence of Candida auris as a global pathogen has been described as a serious global threat by the CDC. It has caused outbreaks in healthcare settings as it is transmissible between patients. The risk factors for candidemia caused by C. auris may be different than candidemia caused by other Candida spp. Methods We performed a multicenter, retrospective case-control study at three hospitals in Brooklyn, New York between 2016 and 2020. Patients with at least one positive blood culture for Candida spp who were started empirically on an antifungal within 24 hours of blood culture positivity were included in the study. Subsequent cases in the same patient were excluded unless separated by at least 90 days from the initial case. Similar variables such as antibiotics and antifungals within the same drug class were compressed into one variable. Variables with a p-value ≤ 0.05 on univariate analysis were entered into a multivariable analysis with a p-value ≤ 0.05 considered to be statistically significant. Results 84 cases of C. auris candidemia and 105 cases of candidemia caused by other Candida spp were included in the analysis. The most common species of other Candida spp was C. glabrata (N=33, 31.7%) followed by C. albicans (N=32, 30.4%). In the multivariable model, the strongest risk factor for C. auris candidemia was prior infection or colonization with C. auris (aOR 17.5; 95% CI, 1.60-192.93; P = 0.019) followed by prior infection or colonization with multidrug-resistant bacteria (aOR 6.97; 95% CI 1.49-32.74, P = 0.014). A history of peripheral vascular disease (PVD) (aOR 7.78; 95% CI 1.34-45.34, P = 0.023), cerebrovascular disease (CVA) (aOR 4.24; 95% CI 1.18-15.20, P = 0.027) and hemiplegia (aOR 6.43; 95% CI 1.19-34.85, P = 0.031) were also statistically significant. These risk factors remained significant analyzing only patients without any history of C. auris. Conclusion These data suggest that in hospitalized patients with candidemia, a history of colonization or infection with C. auris, prior infection or colonization with multidrug-resistant bacteria, as well as a history of PVD, CVA, and hemiplegia are associated with C. auris candidemia. Disclosures Samuel Simon, PharmD, Accelerate Diagnostics (Employee)


Comparison of Initial CXR to CT in Patients (pts) with Hematologic
Background. There is a spectrum of pulmonary disease burden in pts with HEM and PM. There have not been any data comparing the sensitivity and findings of initial Chest X-Ray (CXR) and chest CT in these pts.
Methods. We compared the findings of the initial CXR and CT in all pts with proven or probable PM via EORTC/MSG criteria. We included only pts who had pulmonary symptoms and who had both CXR and CT within 5 days of each other and within seven days of symptom onset or date of culture at MD Anderson Cancer Center from April 2000 and April 2020. We collected data regarding demographics, status of HEM, clinical presentation, frequency and findings of BAL and imaging findings, mold-active prophylaxis and treatment regimens, and mortality. CXR findings were classified as normal or abnormal, and if abnormal sub-classified as mass-like/consolidative, nodular, cavitary or heterogenous/non-specific. CT findings were classified in a similar manner.
Results. We Identified such 39 pts with PM who had both CXR and CT within 5 d. All pts had positive CT. Five pts (13%) had a negative CXR. The majority of pts 28 (72%) were neutropenic (neutrophil count < 500). The most common CXR findings were consolidation or mass-like lesions (56%), followed by patchy, heterogenous or non-specific findings (33%) and nodules (13%). Only 3% had cavitary lesions. Similarly, consolidation or mass-like lesions were the most common finding on CT (69%), followed by nodular lesions with or without ground glass halos (56%). Cavitary lesions and/or reverse halo sign (RHS) were common (31%) on CT. Patients with normal CXR vs those with abnormal CXR were comparable in all clinical parameters we collected. The median survival from time of symptoms onset for all pts was 45 days. There was a trend for lower 42 day mortality in pts with normal CXR (20% vs 47%, P=.253).

Conclusion.
A negative CXR does not preclude PM, especially in neutropenic pts. A CT is recommended for better sensitivity and although there was concordance in CXR with CT findings in some chest abnormalities (mass, consolidation), CT more commonly revealed nodules and signs highly suggestive of PM such as RHS. Although small numbers precluded a robust comparison, it is possible that HEM pts with PM and negative initial CXR have better prognosis, perhaps reflecting a lower burden of pulmonary involvement Disclosures. Dimitrios P. Kontoyiannis, MD, Astellas (Consultant)Cidara Therapeutics (Advisor or Review Panel member)Gilead Sciences (Consultant, Grant/ Research Support, Other Financial or Material Support, Honoraria) Background. Histoplasmosis is a common endemic fungal infection in the Americas, causing significant morbidity and mortality, particularly in immunocompromised patients. Existing diagnostic methods are limited in their sensitivity (especially in pulmonary histoplasmosis) and turnaround time.

A Unique Breath Secondary Metabolite Volatile Signature for the Diagnosis of Histoplasmosis
Methods. We examined prospectively collected breath samples from 84 patients with suspected histoplasmosis 3/2019 -2/2020 at Hospital Roosevelt (HR; Guatemala City, Guatemala, n = 56) and suspected invasive fungal disease 1/2018 -10/2019 at Brigham and Women's Hospital (BWH; Boston, MA, USA, n = 28) using thermal desorption gas chromatography-tandem mass spectrometry (TDU-GC-MS/MS). Patients were evaluated for histoplasmosis and other infections according to the local standard of care -of note, 18/56 patients at HR did not have Histoplasma urine antigen testing.